better url validation regex

oda_api/misc_helpers.py

@@ -5,15 +5,13 @@
 
 regex_url = re.compile(
                 r'^https?://'  # http:// or https://
-                r'(?:(?:[A-Z0-9](?:[A-Z0-9-]{0,61}[A-Z0-9])?\.)+(?:[A-Z]{2,6}\.?|[A-Z0-9-]{2,}\.?)|'  # domain...
-                r'localhost|'  # localhost...
-                r'\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3})'  # ...or ip
+                r'[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*'  # domain...
                 r'(?::\d+)?'  # optional port
-                r'(?:/?|[/?]\S+)$', re.IGNORECASE)
+                r'(?:/[a-z-/]*)?$', re.IGNORECASE)
 
 
 def validate_url(url):
-    return re.match(regex_url, url) is not None
+    return regex_url.match(url) is not None
 
 
 def clean_var_name(s):

tests/test_basic.py

@@ -8,6 +8,7 @@
 import requests
 import oda_api.api
 import oda_api.token
+from oda_api.misc_helpers import validate_url
 
 from cdci_data_analysis.pytest_fixtures import DispatcherJobState
 from conftest import remove_old_token_files, remove_scratch_folders
@@ -229,6 +230,31 @@
     assert disp.url == "https://www.astro.unige.ch/mmoda/dispatch-data"
 
 
+@pytest.mark.parametrize("protocol", [('http://', True), ('https://', True), ('', False)])
+@pytest.mark.parametrize("hostname", [("localhost", True), 
+                                      ("oda-dispatcher", True), 
+                                      ("foo.bar.baz", True), 
+                                      ("1.2.3.4", True), 
+                                      ("", False),
+                                      ("*", False),
+                                      ("foo-", False,),
+                                      ("bar.", False)])
+@pytest.mark.parametrize("port", [("", True), (":8000", True), (':ab', False)])
+@pytest.mark.parametrize("path", [("", True), 
+                                  ("/", True), 
+                                  ("/foo", True), 
+                                  ("/foo/", True), 
+                                  ("/foo/b-ar", True),
+                                  ("/foo/b^ar", False)])
+def test_validate_url(protocol, hostname, port, path):
+    url = f"{protocol[0]}{hostname[0]}{port[0]}{path[0]}"
+    if protocol[1] and hostname[1] and port[1] and path[1]:
+        assert validate_url(url) is True
+    else:
+        assert validate_url(url) is False
+
+
+
 @pytest.mark.parametrize("protocol_url", ['http://', 'https://', ''])
 @pytest.mark.parametrize("init_parameter", ['host', 'url'])
 @pytest.mark.parametrize("protocol_parameter_value", ['http', 'https', '', None, 'not_included'])